-
Notifications
You must be signed in to change notification settings - Fork 2.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
service-catalog cross build test with dind #8734
Conversation
/assign @BenTheElder |
fixed the bad indent on the yaml. |
okay, should be really real this time. where's yaml-fmt when you need it? |
always_run: true | ||
skip_report: false | ||
rerun_command: "/test pull-service-catalog-xbuild" | ||
trigger: "/test( all| pull-service-catalog-xbuild)" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
you probably want to copy the above regex
always_run: true | ||
skip_report: false | ||
rerun_command: "/test pull-service-catalog-xbuild" | ||
trigger: "(?m)^/test( all|pull-service-catalog-xbuild)\\s*" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
lol, /test (all|pull-service-catalog-xbuild)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
how did that space get in there, 😞
0b79da5
to
063a59d
Compare
I have a very powerful case of "the dumb" today. How can I run this check locally? |
yay green. |
- "images-all" | ||
- "svcat-all" | ||
env: | ||
- name: DOCKER_IN_DOCKER_ENABLED |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm going to turn this into a preset soon. Initially we didn't because we wanted to discourage depending on it (I only initially built the support to help us get the rest of our existing jobs onto Prow), but so far it works just fine.
This must be paired with an emptyDir volume for the the graph storage (which is remapped /var/lib/docker
for :reasons:, we should probably unset the option that changes that). Unless things have improved this is necessary to avoid issues with overlays on overlays. Existing docker-in-docker jobs should have this.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What is a preset?
I was basing off of, which has preset-service-account
https://github.com/kubernetes/test-infra/blob/68c68683a066e7b9986113c73a6dcb0f48775ef9/config/jobs/kubernetes-sigs/cluster-api-provider-openstack/cluster-api-provider-openstack-presubmits.yaml
It does not have a emptydir, so maybe it's a bad example, oops.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
searched for emptyDir
and found helm/charts repo
https://github.com/kubernetes/test-infra/blob/1c40d5125cd571010e6949fe0b41065f5973542c/config/jobs/helm/charts/charts.yaml
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
preset is something we are mimicing https://kubernetes.io/docs/concepts/workloads/pods/podpreset/
- args: | ||
- "--repo=sigs.k8s.io/$(REPO_NAME)=$(PULL_REFS)" | ||
- "--root=/go/src/github.com/kubernetes-incubator/$(REPO_NAME)" | ||
- "--timeout=90" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
does this take 90 minutes? if so, does it use many cores / gbs of ram? if it does, we should add a resource request. if it doesn't do any of those things, we should tighten the timeout.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
it's a go build, so it'll use cores we give it. What's the default? 1/2/500millicore?
the two make targets together take about 50minutes on travis, but much less locally, so I'm not sure how to gauge.
This is pretty much the most cpu intensive workload we have. Others take long, but are filled with pauses and inefficiencies. This should be straight compiling.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'll change it to 60, if that sounds okay. What is the default?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
default is infinite I think
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@BenTheElder thanks for help. didn't mean to do a review, oops.
- "images-all" | ||
- "svcat-all" | ||
env: | ||
- name: DOCKER_IN_DOCKER_ENABLED |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What is a preset?
I was basing off of, which has preset-service-account
https://github.com/kubernetes/test-infra/blob/68c68683a066e7b9986113c73a6dcb0f48775ef9/config/jobs/kubernetes-sigs/cluster-api-provider-openstack/cluster-api-provider-openstack-presubmits.yaml
It does not have a emptydir, so maybe it's a bad example, oops.
- args: | ||
- "--repo=sigs.k8s.io/$(REPO_NAME)=$(PULL_REFS)" | ||
- "--root=/go/src/github.com/kubernetes-incubator/$(REPO_NAME)" | ||
- "--timeout=90" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
it's a go build, so it'll use cores we give it. What's the default? 1/2/500millicore?
the two make targets together take about 50minutes on travis, but much less locally, so I'm not sure how to gauge.
This is pretty much the most cpu intensive workload we have. Others take long, but are filled with pauses and inefficiencies. This should be straight compiling.
/lgtm |
/lgtm cancel |
lol, there's a test for it. Why didn't that show up earlier?
Is there a way to run these tests? I'll go search for what's already failed and try to read some. |
I understand now. I'll add that back in. |
from gubernator link, you can find the test command: |
I cannot find the previous snapshot, but I assume you had it before and removed it at some point? |
/lgtm |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: krzyzacy, MHBauer The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
snapshot of what? |
how your initial commit looks like :-) |
I never put it in there, but I mentioned it in the original comment that I saw it and didn't know what it did. |
Does this seem sane now? |
/hold cancel |
@MHBauer: Updated the In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
Yes, you did lgtm it again. I like a little bit more context I suppose. |
lol, lgtm means you are doing good :-) |
@krzyzacy alright, I need some help kubernetes-retired/service-catalog#2212 not sure if this takes longer to take effect, or if we're having more repo permission issues, or how to debug this. |
containers: | ||
- image: gcr.io/k8s-testimages/kubekins-e2e:v20180716-9145034c9-1.10 | ||
args: | ||
- "--repo=sigs.k8s.io/$(REPO_NAME)=$(PULL_REFS)" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this should be github.com/kubernetes-incubator instead of sigs.k8s.io
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
of course it should 🙃
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
followup fix in #8763
I created this by guessing that I need a docker-in-docker enabled instance.
I found the variable DOCKER_IN_DOCKER_ENABLED, and searched for it.
I picked this to copy from
https://github.com/kubernetes/test-infra/blob/68c68683a066e7b9986113c73a6dcb0f48775ef9/config/jobs/kubernetes-sigs/cluster-api-provider-openstack/cluster-api-provider-openstack-presubmits.yaml
because it's in the new location for config, so I figure that means someone has looked at it.
I didn't totally copy it, because I don't know what some of the args are, such as:
and
- "--upload=gs://kubernetes-jenkins/pr-logs"
I also don't know how much memory it needs, so I dropped the 6G requirement. I do not know what the default is, but I'm going to hope it's enough for now.